Automated Query Relaxation Mechanism for QoS-Aware Service Provisioning

Abstract

Offering services maintaining the requested QoS-levels is the main concern of service discovery approaches. The efficiency of such approaches can be measured often in terms of execution time. Most of the existing works focus on the low execution time but do not consider the provision of closer alternatives in case of “unsuccessful search” . Search method becomes unsuccessful if the query fails to discover services having necessary QoS metrics with desired levels. Available alternatives with a minimum compromise in case of “unsuccessful search” may enhance the efficiency. Method of query relaxation is a well-accepted approach for enhancing customer satisfaction. All the existing works on query relaxation concentrate on relaxing strategy while compromising execution time. Moreover, often the inputs from users regarding preference on QoS metrics are needed for offering a solution. These intermediate interventions increase execution time, and the solution highly depends on the domain knowledge of the consumers. Here, we have proposed an approach, capable of offering a set of alternative solution through query relaxation in a fully automated way without compromising in terms of execution time. A lattice-based meta-model with a subsequent discovery mechanism is presented that relaxes the query in terms of QoSs for providing possible alternatives. Moreover, the solution exhibits well with execution time complexity of (O(klogn)) that is comparable to the best solution in the service discovery domain.